GTX 480 vs. HD 5870, 8x AA Performance Analysis, Part 3

Test Configuration

Test Configuration – Hardware

  • Intel Core i7 920 reference 2.66 GHz and overclocked to 3.8 GHz); Turbo (21X multiplier for 3.97 GHz of a single core) is on.
  • Gigabyte EX58-UD3R (Intel X58 chipset, PCIe 2.0 specification; CrossFire/SLI 16x+16x).
  • 6 GB OCZ DDR3 PC 18000 Kingston RAM (3×2 GB, tri-channel at PC 16000 speeds; 2×2 GB supplied by Kingston)
  • NVIDIA GTX 480, reference design (supplied by NVIDIA formerly under NDA)
  • ATI Radeon HD 5870 (2GB, reference clocks) by Diamond
  • Onboard Realtek Audio
  • Two identical 250 GB Seagate Barracuda 7200.10 hard drives configured and set up identically from drive image; one partition for NVIDIA GeForce drivers and one for ATI Catalyst drivers
  • SilentPro 600 M power supply unit supplied by Cooler Master
  • Cooler Master Gladiator 600 Case supplied by Cooler Master
  • Noctua NH-U12P SE2 CPU cooler supplied by Noctua
  • Four Case fans by Cooler Master and one Noctua NF-P14 FLX
  • Philips DVD SATA writer
  • HP LP3065 2560×1600 thirty inch LCD

Test Configuration – Software

  • ATi Catalyst 10-3a; highest quality mip-mapping set in the driver, Catalyst AI set to “Standard”
  • NVIDIA GeForce 197.41 WHQL drivers for GTX 4×0; High Quality
  • Windows 7 64-bit; very latest updates
  • DirectX February 2010
  • All games are patched to their latest versions.
  • vsync is off in the control panel and is never set in-game.
  • 4xAA enabled in all games and “forced” in Catalyst Control Center for UT3; all in-game settings at “maximum” or “ultra” with 16xAF always applied if possible; 16xAF forced in control panel Crysis.
  • All results show average, minimum and maximum frame rates except as noted.
  • Highest quality sound (stereo) used in all games.
  • Windows 7 64, all DX10 titles were run under DX10 render paths; DX11 titles under DX 11 render paths except for Dirt 2 demo in DX9c.

The Benchmarks

  • Batman Arkham Asylum
  • Just Cause 2
  • Enemy Territory: Quake Wars
  • Crysis
  • Far Cry 2
  • World in Conflict
  • X3:Terran Conflict
  • Dirt 2
  • Left4Dead
  • Lost Planet
  • Unreal Tournament 3
  • Resident Evil 5
  • PT Boats
  • ARMA2
  • H.A.W.X.
  • Battleforge
  • Heaven 1.0 (Unigine)
  • Heaven 2.0 (Unigine)

We dumped Vantage and added three more games – an old favorite, Enemy Territories: Quake Wars and two new games, Batman Arkham Asylum and Just Cause 2. We also added Heaven 2.0 to Part Two as a second synthetic benchmark because it uses extreme shaders and it may well portend the future of games with DX11′s most noticeable feature over DX10, tessellation. In fact, the Unigine Engine will be the basis for at least two new DX11 games this year, one of which is Primal Carnage – a dinosaur game.  Join us now as we explore the performance hit of 8xAA over 4xAA with both of our current competing top single-GPU video cards. Let’s move on to our 16 game benchmark suite as well as the two Unigine benchmarks as we explore the relative performance of the GTX 480 versus HD 5870 as they are each tested at 8xAA to see the relative framerate hit on each card.

«»

Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

apoppin

Founder and Senior Editor of ABT.

5 Responses

  1. hmmm.... says:

    Please take into consideration that nVidia uses a different version of AntiAliasing starting 8x and up, therefore comparisons are henceforth limited at best. Sadly I don’t have a direct link right now, but please take it into consideration before drawing (final) conclusions.

  2. apoppin says:

    We took special care to make sure that identical AA settings were applied in all of our benchmarks including for Crysis. We even noted that in the full retail game, Just Cause 2, that we observed the benchmark results showed the Radeon was running at 8xCSAA while the GeForce was 8xAA.

    However, we have since learned from AMD that the benchmark results are wrongly identifying 8xMSAA as CSAA. The Radeon is actually running 8xMSAA and this minor issue will be addressed in a future patch.

    Everything we test is “apples to apple” unless it is specified in the review.

  3. Nice article man. Cheers

  4. Ahhroorah says:

    Amazing how you calculated the % differences and included them in the chart for 4x vs 8x AA! One of the best reviews ever.

Leave a Reply

Your email address will not be published.

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>